The Demand for Health: Theoretical Underpinnings and Empirical Results | PART 1 |
In the fall of 1966, Gary S. Becker was a member of a National Bureau of Economic Research (NBER) staff reading committee that reviewed a paper titled āThe Production of Health: An Exploratory Study,ā by Richard Auster, Irving Leveson, and Deborah Sarachek. Garyās main comment on the paper was that it ignored that what people demand when they purchase medical care services are not these services per se but rather good health. The latter item enters the utility functions of consumers, and medical care is only one of many inputs into its production. He proceeded to specify a demand function for health whose arguments included the prices of health inputs, the efficiency of the production process as reflected by the number of years of formal schooling completed by the consumer, and income or, more precisely, the exogenous components of income.
When Gary wrote his review, I had just entered my third year in the PhD program in economics at Columbia University. He was a professor of economics at that university as well as an NBER research associate. I had completed all the courses and examinations and was searching for a dissertation topic. I also was working as a research assistant to Victor R. Fuchs at the NBER. Victor had encouraged me to select a topic in health economics for my dissertation. I had consulted Gary about a topic, but he had not worked in health economics and did not have any suggestions. His review changed all that. After Gary finished it, he gave it to me and said that it contained an idea for a PhD dissertation. Originally, it was supposed to be a study of the effects of education on health, but along the way he encouraged (some might say demanded) me to broaden it into a theoretical and empirical analysis of the demand for health.
It is not surprising that Gary emphasized the difference between health as an output that enters the utility function and medical care as one of a number of inputs into its production in his review. After all, he had published his seminal paper titled āA Theory of the Allocation of Timeā in the Economic Journal the previous year (Becker 1965). In that paper, he developed the household production function model of consumer behavior by drawing a distinction between fundamental objects of choice (called commodities) that enter the utility function and market goods and services. Consumers produce commodities with inputs of market goods and services and their own time. Because goods and services are inputs into the production of commodities, the demand for these goods and services is a derived demand for a factor of production. Although most of the applications in his paper pertain to extensions of labor supply theory, Gary could easily have titled it āThe Household Production Approach to Consumer Behaviorā or āOn the New Theory of Consumer Behaviorā because the applications extended far beyond labor supply. Indeed, he published a paper with the latter title in the Swedish Journal of Economics in 1973 (Michael and Becker 1973).
In my dissertation (Grossman 1970) and the publications that resulted from it (Grossman 1972a, 1972b), I use the household production function approach as one of two building blocks to construct a model of the demand for health. I assume that consumers demand health, defined broadly to include illness-free days in a given year and life expectancy, and produce it with inputs of medical care services, diet, other market goods and services, and their own time. Hence, the demand for medical care and other health inputs is derived from the basic demand for health.
My second building block, also due to Gary, is the theory of investment in human capital (for example, Becker 1964). Health, like knowledge, is a durable capital stock; and both may be viewed as components of the stock of human capital. Consumers have incentives to invest in this stock in the present because it increases their earnings in the future. Indeed, in his Economic Journal paper, Gary pointed out that investment in human capital is a prominent use of a portion of the time allocated to nonmarket or household production. I proceeded to pursue a distinction between the returns to an investment in knowledge and the returns to an investment in health that he suggested to me. To be specific, investments in knowledge raise wage rates and investments in health raise the total amount of time available for market and household production in a given year and prolong length of life.1
The first paper in this section contains most of the theory in my dissertation. It appeared in the 1972 volume of the Journal of Political Economy with the title āOn the Concept of Health Capital and the Demand for Health.ā It is my first publication of original material and is my most widely cited one. In the paper, I first develop a general model of the demand for health, which contains both an investment motive and a consumption motive for increasing the stock of health. I then focus on a pure investment model in which the consumption benefits of health are small enough to be ignored. I do so to contrast health capital with other forms of human capital and because the pure investment model generates powerful predictions from simple analyses. Moreover, the consumption aspects of the demand for health can be incorporated into empirical estimation without much loss in generality.
The other paper in this section is my contribution to Volume 1A of the Handbook of Health Economics, published by Elsevier (Grossman 2000). I include it as a substitute for my 1972 monograph titled The Demand for Health: A Theoretical and Empirical Investigation, published by the NBER.2 The monograph and the paper spell out the differences between a pure investment model of the demand for health and a pure consumption model in which the investment returns are small enough to be ignored. The paper also summarizes the main empirical results in the 1972 monograph. In addition, because the paper was published almost three decades after the monograph, some extensions of the theoretical and empirical work in the monograph and some criticisms of the framework as of the late 1990s are discussed.
I would be remiss if I did not indicate the fate of the paper by Auster, Leveson, and Sarachek that led Gary Becker to suggest the topic of my PhD dissertation to me. I am delighted to report that it was published in an early volume of the Journal of Human Resources (Auster, Leveson, and Sarachek 1969). The authors concluded that the rate of return to investing in health by increasing education far exceeded the rate of return to investment in health by increasing medical care. That conclusion was responsible, in part, for the subsequent literature that has investigated whether more schooling causes better health. That issue serves as the subject of part 2 of this book.
NOTES
1. In the introductions and afterwords to the first two parts of this book, I refer to forms of human capital other than health capital as knowledge capital, which is most commonly measured at the empirical level by completed years of formal schooling.
2. I am very pleased that Columbia University Press, which distributed the monograph for the NBER, is reissuing the monograph as a companion to this book.
REFERENCES
Auster, Richard, Irving Leveson, and Deborah S...